One of the most important components of educational research over the past two decades has been the emergence of cooperative learning as a dominant learning paradigm. This paradigm emphasizes learning as an active process best facilitated by interaction with peers, teachers, and other learning resources \cite{Slavin91}. The focus of a cooperative setting is on group directed, rather than individually directed effort, which presumes positive interdependencies among the members of the group. Learners work together to discuss learning problems and share responsibility for the learning of the group, as well as for each individual member.
There is general consensus that individuals working in groups produce higher achievement than those working alone, due to the increase in cognitive resources. Other advantages of collaboration in education include co-construction, social facilitation, verbalization, conflict, observation, increased motivation and self-esteem \cite {Slavin91,Issroff96}.
A collaborative setting may, however, also include negative interdependencies amongst its members. Cooperation in this case may mean competition, where one student's gain is the other's loss, or be differently balanced for each member of a group: some learners do not learn well in collaborative settings, while others show increased productivity \cite{Johnson85}. Collaborative learning doesn't ensure that everyone will achieve the same level or type of mastery. Simply placing learners together does not mean that they will collaborate. As shown through classroom practice, collaborative learning often degenerates into working in groups with highly non-uniform participation and effort by group members.
As long as students need help in developing skills for cooperative work, the development of collaborative learning environments requires a framework which emphasizes structure and clearly defined goals, so to encourage individual responsibility in such a way that students are concerned about the performance of the whole group, not just their own \cite{Johnson85}. Computers certainly afford opportunities for improved control, explicit support for differentiated roles, mechanisms for planning group activities and providing feedback on groups' progress towards a goal. A number of projects in the field of Computer Supported Cooperative Learning (CSCL) study the use of computers in augmenting and supporting group work \cite {Johnson86}. A central theme in many of these studies is the notion of verbal interaction between pairs or groups of computer users \cite {Steiner95,Steiner94,Steiner94a}.
The idea of computer-supported cooperative learning is recently being extended through the emergence of telepresence. The integration of virtual reality and high-speed networking \cite{Leigh96a} expands physical boundaries and opens up new possibilities for collaboration beyond the classroom. Through the use of avatars, or virtual personas, geographically separated learners perceive simultaneous presence in a virtual environment \cite{Dede95}. The ability to connect with learners at distant locations, enhanced by visual, gestural, and verbal interaction is key to the development of unique educational experiences. Some of the advantages here include the ability for referential communication (to ``put myself in your shoes''), distributed cognition amongst learners, as well as issues of verbal interaction and cultural awareness.